Journal of Fungi

Article Genetic Manipulation of the Brassicaceae Thecaphora thlaspeos

Lesley Plücker †, Kristin Bösch †, Lea Geißl, Philipp Hoffmann and Vera Göhre *

Institute of Microbiology, Cluster of Excellence in Plant Sciences, Heinrich-Heine University, Building 26.24.01, Universitätsstr.1, 40205 Düsseldorf, Germany; [email protected] (L.P.); [email protected] (K.B.); [email protected] (L.G.); [email protected] (P.H.) * Correspondence: [email protected]; Tel.: +49-211-811-1529 † These authors contribute equally to this work.

Abstract: Investigation of plant–microbe interactions greatly benefit from genetically tractable part- ners to address, molecularly, the virulence and defense mechanisms. The smut fungus maydis is a model pathogen in that sense: efficient homologous recombination and a small genome allow targeted modification. On the host side, is limiting with regard to rapid genetic alterations. By contrast, the model plant is an excellent model with a vast amount of information and techniques as well as genetic resources. Here, we present a transformation protocol for the Brassicaceae smut fungus Thecaphora thlaspeos. Using the well-established methodology of protoplast transformation, we generated the first reporter strains expressing fluorescent proteins to follow mating. As a proof-of-principle for homologous recombination, we deleted the pheromone receptor pra1. As expected, this mutant cannot mate. Further analysis will contribute to our understanding of the role of mating for biology in this novel model fungus. From now on, the genetic manipulation of T. thlaspeos, which is able to colonize the model plant A. thaliana, provides us with a pathosystem in which both partners are genetically amenable to study smut infection biology.

  Keywords: mating; pheromone receptor; homologous recombination; infection; smut; transforma- tion; protoplast Citation: Plücker, L.; Bösch, K.; Geißl, L.; Hoffmann, P.; Göhre, V. Genetic Manipulation of the Brassicaceae Smut Fungus Thecaphora 1. Introduction thlaspeos. J. Fungi 2021, 7, 38. https://doi.org/10.3390/jof7010038 Smut fungi are important pathogens causing economic losses in crops such as , , maize, and potato [1]. The dimorphic lifecycle of grass smut fungi is comprised Received: 16 November 2020 of a yeast form [2] that, in contrast to many other biotrophic fungi, can be cultured, and Accepted: 7 January 2021 is amenable to genetic manipulation [3]. In combination with the efficient homologous Published: 9 January 2021 recombination, this has turned Ustilago maydis, the maize smut fungus [4], into an important model organism for fungal and infection biology [5,6]. Publisher’s Note: MDPI stays neu- U. maydis starts the infection by mating, resulting in the morphological switch to tral with regard to jurisdictional clai- the infectious filamentous form. Mating in smut fungi is controlled by two mating loci. ms in published maps and institutio- The a locus encodes for pheromones (mfa) and pheromone receptors (pra) that mediate nal affiliations. recognition of compatible mating partners and trigger cell fusion [7]. The b locus contains two (bW, bE) encoding for subunits of a heterodimeric, homeodomain transcription factor. When an active transcription factor is assembled from different alleles in the common cytosol after plasmogamy, filamentous growth is initiated and thereby the fungus switches Copyright: © 2021 by the authors. Li- censee MDPI, Basel, Switzerland. from saprophytic to pathogenic growth [8]. Notably, the infectious filaments are arrested This article is an open access article in the cell cycle until successful penetration of the host plant [9]. This mating system is distributed under the terms and con- widely conserved in grass smut fungi and allows for genetic exchange between the mating ditions of the Creative Commons At- partners [10]. tribution (CC BY) license (https:// In contrast to the well-characterized grass smut fungi infecting important monocot creativecommons.org/licenses/by/ crop plants with U. maydis at the forefront, little is known about smut fungi infecting 4.0/). dicot plants. A reemerging example is Microbotryum that is regaining attention today [11].

J. Fungi 2021, 7, 38. https://doi.org/10.3390/jof7010038 https://www.mdpi.com/journal/jof J. Fungi 2021, 7, 38 2 of 14

By contrast, the Thecaphora-clade [12], with agronomically relevant members such as T. solani [13] or T. frezii [14] infecting potato and peanut, respectively, remains largely elusive. One member, the Brassicaceae smut fungus T. thlaspeos, infects several Arabis species [12] and can colonize Arabidopsis thaliana [15], making it a good system to study smut infection in model plants. Interestingly, germinating teliospores of T. thlaspeos directly give rise to an infectious filament, and no saprophytic phase is known. However, prolonged cultivation leads to fungal proliferation as filamentous, haploid cultures. From such cultures, two filamentous haploid T. thlaspeos strains of compatible mating types, LF1 and LF2, could be isolated [15]. The unique germination pattern and the emergence of haploid filaments in culture raise questions about mating and meiosis in T. thlaspeos. The genome of T. thlaspeos was recently sequenced and annotated. With 20.5 Mb, 6239 models, and a low repeat content, it is a typical smut fungal genome [16]. Notably, we could identify the mating loci a and b in this genome. In comparison to the grass smut fungi, the a locus of T. thlaspeos is strongly rearranged, and aligns well to the biocontrol yeast Anthracocystis flocculosa (formerly Pseudozyma flocculosa). It still contains one copy of the pheromone receptor pra1 and pheromone mfa1, or pra2 and mfa2 in T. thlaspeos LF1 and LF2, respectively [16]. The b locus is conserved with a bi-directional promoter in between the two genes, and we have previously shown that the heterodimer formation of bE and bW isconserved in T. thlaspeos [15]. Conservation of the mating genes throughout evolution suggests that T. thlaspeos still uses mating, e.g., for exchange of genetic material. However, the infectious lifecycle so far has not revealed a stage where mating is required. Therefore, we aimed for a more detailed understanding of the role of the T. thlaspeos mating genes. To study the mating process in T. thlaspeos, genetic manipulation is essential. For example, reporter lines enable life-cell imaging, and deletion mutants can give insight into mutant phenotypes. Several transformation methods have been developed for smut fungi. Protoplast-mediated transformation is used for the grass smuts U. maydis [17], Sporisorium reilianum [18], U. hordei [19], U. esculenta [20], and U. bromivora [21], as well as the filamentous Basidiomycete Serendipita indica (formerly Piriformospora indica)[22], and the biocontrol yeast An. flocculosa [23]. Agrobacterium-mediated transformation is used for S. scitamineum in combination with a CRISPR-Cas9 system [24], and also for bringing larger fragments into U. hordei [25] and for gene tagging in U. maydis [26]. Here, we have established a protoplast-based PEG-mediated transformation system and generated targeted deletion mutants to investigate the mating process in T. thlaspeos.

2. Materials and Methods 2.1. Fungal Culture Conditions T. thlaspeos LF1 and LF2 haploid strains [15] from our own collection were used in this study. Both were grown in YEPS light liquid medium (1% yeast extract w/v, 0.4% w/v Bacto TM-Peptone, and 0.4% w/v sucrose) or YEPS light solid medium with 0.6% w/v plant agar or 1% w/v phytagel at 18 ◦C. Cryostocks of T. thlaspeos were generated by mixing exponentially growing cultures with 30% glycerol in growth medium followed by immediate freezing at −80 ◦C. Filamentous cultures were started by resuspending the glycerol stock in growth medium or plating the cells on solid medium as described above.

2.2. Plasmid Construction All plasmids used in this study were generated with the Golden Gate Cloning tech- nique (Protocol S1) as described [27]. The hpt-egfp and hpt-mcherry sequences were codon- optimized for U. maydis. Promoter and terminator sequences from T. thlaspeos or U. maydis were amplified via PCR from genomic DNA.

2.3. Protoplasting

T. thlaspeos cultures were grown in YEPS light medium to an OD600 of 0.5–0.8 for 3 to 4 days. Fungal mycelium was collected using cell strainers with 40 µm mesh size (VWR TM Darmstadt, Germany) and washed with protoplasting buffer (0.1 M sodium citrate, 0.01 M J. Fungi 2021, 7, 38 3 of 14

EDTA 1.2 M MgSO4, and pH 5.8) to remove residual culture medium. T. thlaspeos tissue was resuspended in 9 mL protoplasting buffer, supplemented with 10 mg/mL Yatalase (Takara Bio, Kusatsu, Japan) and 20 mg/mL Glucanex (Sigma-Aldrich, St. Luis, MI, USA) per 100 mL cell culture, and incubated for 30–60 min at room temperature. Protoplast formation was controlled microscopically. When protoplasting was finished, protoplasting buffer was added to a total volume of 24 mL. Aliquots of 6 mL crude protoplast solution were overlayed with 5 mL trapping buffer (0.6 M sorbitol, 0.1 M Tris/HCl pH 7.0) and centrifuged at 4863× g (5000 rpm) in a swing out rotor at 4 ◦C for 15 min. The interphase was collected from all tubes and diluted with 2 volumes of ice-cold STC buffer (0.01 M Tris/HCl pH 7.5, 0.1 M CaCl2, and 1.0 M sorbitol). Protoplasts were pelleted at 4863× g (5000 rpm) in a swing out rotor at 4 ◦C for 10 min and resuspended in 500 µL ice-cold STC buffer. 100 µL protoplast aliquots were used for transformation immediately. A bullet point version of the protocol is available with the supplementary files (Protocol S1).

2.4. Transformation Transformation of T. thlaspeos protoplasts were carried out as described for U. may- dis [28] with slight modifications. Transformation reactions were spread on YMPG-Reg (0.3% w/v Yeast extract, 0.3% w/v malt extract, 0.5% w/v Bacto-Peptone, 1% w/v glucose, 1 M sucrose, 0.6% w/v plant agar, Duchefa Biochemie, Haarlem, Netherlands) medium and incubated at 18 ◦C until colonies appeared (1–2 months). Selection was carried out on 10 µg/mL Hygromycin B (Roth, Karlsruhe Germany) following the layered-plate strategy used for U. maydis [28]. In this setup, hygromycin was provided in the bottom layer so that it took time to diffuse to the top before the selection took place. This gave protoplasts time to regenerate and express the resistance gene [28]. Colonies were singled-out on YEPS light solid medium supplemented with 10 µg/mL hygromycin. Single colonies were then used to inoculate YEPS light liquid cultures for molecular analysis.

2.5. Molecular Analysis of Transformants Successful integration of the constructs were determined by PCR and Southern Blot analysis [28], and eGfp or mCherry fluorescence was used as a rapid indicator for expres- sion of the constructs. Genomic DNA of T. thlaspeos was extracted using the NEB Monarch Genomic DNA Purification Kit (New England Biolabs, Frankfurt, Germany).

2.6. Mating Assay For confrontation assays, liquid cultures of T. thlaspeos strains were spotted on YEPS light solid medium in close proximity and allowed to grow towards each other. When the hyphae of both strains were close enough to appear in the display window of the microscope, a time lapse experiment was conducted to monitor mating for 24–72 h. For liquid mating assays, strains of compatible mating types were mixed in YEPS light liquid medium in equal amounts and incubated at 18 ◦C and 200 rpm. Medium was exchanged twice a week. After 8–14 days, plasmogamy was observed microscopically via eGfp and mCherry fluorescence.

2.7. Microscopy Fluorescence microscopy as well time lapse experiments were performed on a Zeiss Axio Immager M1 according to [29]. Microscope control, image acquisition, and processing were done with the software package Meta-Morph (version 7; Molecular Devices).

3. Results 3.1. Protoplast—Generation from Filamentous T. thlaspeos Cultures Protoplast-mediated transformation is well-established for fungi [30]. The protocol for U. maydis [17] was successfully adapted for other smut fungi and therefore was also the starting point for T. thlaspeos protoplast generation. Critical factors besides cultivation conditions are the enzyme mixture, the buffer composition, and the osmotic stabilizer. J. Fungi 2021, 7, x FOR PEER REVIEW 5 of 16

3. Results 3.1. Protoplast—Generation From Filamentous T. thlaspeos Cultures Protoplast-mediated transformation is well-established for fungi [30]. The protocol for U. maydis [17] was successfully adapted for other smut fungi and therefore was also J. Fungi 2021, 7, 38 4 of 14 the starting point for T. thlaspeos protoplast generation. Critical factors besides cultivation conditions are the enzyme mixture, the buffer composition, and the osmotic stabilizer. First, we compared different enzyme mixtures. Glucanex, a mix of lysing enzymes from TrichodermaFirst, we compared harzianum different including enzyme beta-1,3-glucanase mixtures. Glucanex, activity, a works mix of well lysing for enzymesU. maydis from and theTrichoderma filamentous harzianum Basidiomyceteincluding S. beta-1,3-glucanase indica [22]. Yatalase activity, comprises works of well a mix for U.of maydislysing enzymesand the filamentousfrom Corynebacterium, Basidiomycete includingS. indica chit[22inase-,]. Yatalase chitobiase-, comprises and beta-1,3-glucanase of a mix of lysing activity,enzymes for from cell Corynebacterium, wall lysis of filamentous including fungi. chitinase-, In combination chitobiase-, with and Glucanex, beta-1,3-glucanase it is used foractivity, U. bromivora for cell wall[21] lysisand Agrocybe of filamentous aegerita fungi. [31], or In supplemented combination with with Glucanex, chitinase itfor is Aspergillusused for U. niger bromivora [32]. Novozyme[21] and Agrocybe 234, which aegerita worked[31], orvery supplemented well for U. withmaydis chitinase [28], is forno longerAspergillus commercially niger [32]. available, Novozyme so we 234, did which not include worked it very in our well study. for U. In maydispilot studies,[28], is we no comparedlonger commercially enzyme and available, buffer combinations so we did not of includepublished it in protopla our study.sting In protocols pilot studies, and foundwe compared that the enzyme combination and buffer of Glucanex combinations and ofYatalase published efficiently protoplasting protoplasts protocols the and T. thlaspeosfound that filaments the combination (Table S1). of Glucanex and Yatalase efficiently protoplasts the T. thlaspeos filamentsTherefore, (Table to S1). first optimize the osmotic stabilizer, we used this enzyme mix in the U. maydisTherefore, protoplasting to first buffer. optimize Typical the osmotic osmotic stabilizer, stabilizers we are used inorganic this enzyme salts, mix sugars, in the orU. sugarmaydis alcoholsprotoplasting [33]. For buffer. example, Typical sorbitol osmotic is used stabilizers for U. maydis are inorganic, and sucrose salts, sugars,for U. esculenta or sugar. Thus,alcohols we [ 33tested]. For sorbitol example, and sorbitol sucrose, is used as forwellU. as maydis MgSO, and4, which sucrose is for usedU. esculentafrequently. Thus, in combinationwe tested sorbitol with andYatalase. sucrose, Most as well protoplasts as MgSO 4were, which obtained is used using frequently MgSO in4 combination (Figure 1). Notably,with Yatalase. sorbitol Most and protoplasts sucrose inhibited were obtained cell wall using lysis, MgSO confirming4 (Figure1 ).early Notably, observations sorbitol [34,35].and sucrose Subsequently, inhibited cellwe wall tested lysis, various confirming commonly early observations used protoplasting [34,35]. Subsequently, buffers, in we tested various commonly used protoplasting buffers, in combination with MgSO as combination with MgSO4 as the osmotic stabilizer. There were no significant differences4 betweenthe osmotic the stabilizer.four tested There buffers were (Table no significant 1), but we differences observed betweentendencies the that four citrate tested buffers buffers work(Table better1), but for we fast observed growing tendencies cultures that that citrate were bufferssub-cultured work betterbi-weekly for fast (Figure growing S1). Ultimately,cultures that we were decided sub-cultured on the 0.1 bi-weekly M citrate (Figure buffer S1). and Ultimately, 0.01 M EDTA, we decided which on is thealso 0.1 the M An. flocculosa buffercitrate used buffer for and An. 0.01 flocculosa M EDTA,, the whichclosest is homolog also the bufferof T. thlaspeos, used for and a bi-weekly, the splitting closest homolog of T. thlaspeos, and a bi-weekly splitting rhythm. rhythm.

Figure 1. Identification ofFigure an osmotic 1. Identification stabilizer. Thecaphora of an osmotic thlaspeos stabilizer.LF1 cultureThecaphora was thlaspeos grown LF1 to an culture OD600 was= 0.4–0.8. grown Toto an optimize protoplasting of T.OD thlaspeos600 = 0.4–0.8.hyphae To by optimize Yatalase protoplasting and Glucanex, of the T. osmotic thlaspeos stabilizers hyphae by MgSO Yatalase4, sorbitol, and Glucanex, and sucrose the osmotic stabilizers MgSO4, sorbitol, and sucrose were tested. With MgSO4 as osmotic stabilizer, all were tested. With MgSO4 as osmotic stabilizer, all filaments were digested; while in sorbitol and sucrose, no protoplasts were obtained. Black arrowheads:filaments filaments; were digested; white arrowheads: while in sorbitol protoplast; and sucrose, scale bar: no protoplasts 50 µm. were obtained. Black arrowheads: filaments; white arrowheads: protoplast; scale bar: 50 µm. An advantage of MgSO4 as osmotic stabilizer is that the majority of intact protoplasts Tablein the 1. presence Optimizing of MgSOthe protoplasting4 have large buffer. vacuoles To iden [38tify], whichthe optimal enables osmotic collection stabilizer, and fungal purifica- hyphaetion, and were floating filtered in and a trappingincubated buffer in 0.02 [M36 ].citrate Intact buffer, protoplasts supplemented accumulate with different in a sharp osmotic band stabilizersat the interphase, and 10 mg/mL and debris Yatalase pellet + 20 atmg/mL the bottom Glucanex, (Figure for 602). min After at RT. washing Protoplasting with STC worked buffer, 8 onlyup to if 10MgSOprotoplasts/g4 was used as freshosmotic weight stabilizer. can To be optimize recovered. the buffer for the use of MgSO4, hyphae were filteredUpon determinationand incubated in ofdifferent the optimal buffers, buffer supplemented and osmotic with 1.2 stabilizer, M MgSO4 weand reevaluated10 mg/mL Yatalasethe composition + 20 mg/mL and Glucanex, concentration for 60 min of at the RT. enzyme There was mix no (Figure significant3). As difference the pilot between study the had indicatedindicated, buffers, for efficient but a tendency degradation toward ofs higher the T. yields thlaspeos with cellcitrate wall, buffers. the combined activity of Yatalse and GlucanexBuffer is necessary. As expected,Osmotic individually Stabilizer theProtoplast enzymes Yield/g are poorly FW active,× 107 leaving filaments behind (Figure3A). This emphasizes the importance of testing various lysing enzymes, alone and in combination, in different buffers to find a mix suitable for the organism of choice and its individual cell wall composition [30].

Finally, we aimed to decrease the enzyme concentrations to save costs. However, lowering the concentration to half resulted in incomplete digestion of the fungal cell wall after 30 min (Figure3B). Since it was described earlier that a shorter incubation time is J. Fungi 2021, 7, 38 5 of 14

preferable, compared to a low enzyme concentration, regarding protoplast viability [39,40], J. Fungi 2021, 7, x FOR PEER REVIEWwe did not reduce the enzyme concentrations. 6 of 16

Table 1. Optimizing the protoplasting buffer. To identify the optimal osmotic stabilizer, fungal hyphae were filtered and incubated in 0.02 M citrate buffer, supplemented with different osmotic stabilizers and 10 mg/mL Yatalase + 20 mg/mL

Glucanex, for 60 min at RT. Protoplasting worked only ifOptimizing MgSO4 was usedthe Osmotic as osmotic Stabilizer stabilizer. To optimize the buffer for the use of MgSO4, hyphae were filtered and incubated in different buffers,0.4 M supplementedsucrose [20] with 1.2no M MgSOprotoplasts4 and 10 mg/mL Yatalase + 20 mg/mL Glucanex,0.02M for citrate, 60 min atpH RT. 5.8 There [28] was no significant1.2 M MgSO difference4 [36] between the indicated5.52 ± 1.22 buffers, but a tendency towards higher yields with citrate buffers. 1.0 M sorbitol [28] no protoplasts Optimizing the buffer composition Buffer Osmotic Stabilizer Protoplast Yield/g FW × 107 0.1M citrate, 0.01 M EDTA, pH 5.8 [23] 7.46 ± 2.02 Optimizing the Osmotic Stabilizer 0.02M citrate, pH 5.8 [28] 7.49 ± 1.51 0.4 M sucrose [20]1.2 M MgSO4 [36] no protoplasts 0.02M MES, pH 5.8 [21] 5.28 ± 0.66 0.02 M citrate, pH 5.8 [28] 1.2 M MgSO4 [36] 5.52 ± 1.22 0.01 M phosphate,1.0 pH M sorbitol5.8 [37] [ 28] no protoplasts5.01 ± 0.51 Optimizing the buffer composition 0.1 M citrate, 0.01 M EDTA, pH 5.8 [23An] advantage of MgSO4 as osmotic stabilizer is that the majority7.46 ± of2.02 intact protoplasts 0.02 M citrate, pH 5.8 [28] 7.49 ± 1.51 in the presence of MgSO1.2 M MgSO4 have[ 36large] vacuoles [38], which enables collection and 0.02 M MES, pH 5.8 [21]4 5.28 ± 0.66 purification, and floating in a trapping buffer [36]. Intact protoplasts accumulate in a sharp 0.01 M phosphate, pH 5.8 [37] 5.01 ± 0.51 band at the interphase, and debris pellet at the bottom (Figure 2). After washing with STC buffer, up to 108 protoplasts/g fresh weight can be recovered.

FigureFigure 2.2. ProtoplastingProtoplasting filamentousfilamentous T.T. thlaspeosthlaspeos cultures.cultures. 1–21–2 gg freshfresh weightweight ofof anan exponentiallyexponentially growinggrowing cultureculture ((AA)) waswas harvested by filtration and protoplasted with Glucanex and Yatalase for 30–60 min. To purify the crude protoplasts (B), harvested by filtration and protoplasted with Glucanex and Yatalase for 30–60 min. To purify the crude protoplasts (B), they are floated in a gradient (C). Filaments and debris are found in the pellet, the protoplasts can be collected from the they are floated in a gradient (C). Filaments and debris are found in the pellet, the protoplasts can be collected from the interphase (marked with a white box). (D) An efficient protoplasting reaction results in up to 108 protoplasts/g fresh 8 interphaseweight. Scale (marked bar: 10 with µm. a white box). (D) An efficient protoplasting reaction results in up to 10 protoplasts/g fresh weight. Scale bar: 10 µm. Upon determination of the optimal buffer and osmotic stabilizer, we reevaluated the In contrast to other applications of protoplasts, for genetic manipulation, the proto- composition and concentration of the enzyme mix (Figure 3). As the pilot study had plasts have to be viable and able to regenerate their cell wall. Thus, we next investigated indicated, for efficient degradation of the T. thlaspeos cell wall, the combined activity of the influence of the osmotic stabilizer on regeneration of the protoplasts after transforma- Yatalse and Glucanex is necessary. As expected, individually the enzymes are poorly tion. This allowed us to separate optimization of both steps. Using protoplasts generated active, leaving filaments behind (Figure 3A). This emphasizes the importance of testing in the presence of MgSO , we assessed regeneration media containing different osmotic various lysing enzymes,4 alone and in combination, in different buffers to find a mix stabilizers such as sucrose, glucose, sorbitol, and KCl (Figure4). MgSO was excluded due suitable for the organism of choice and its individual cell wall composition4 [30]. to incompatibility with the gelling agent. Maximal regeneration was obtained with 1 M sucrose as osmotic stabilizer, followed by sorbitol and glucose, while 1 M KCl completely inhibited regeneration and fungal growth (Figure4A). In support, T. thlaspeos LF1 cell cultures do not grow on 1 M KCl, indicating that it is toxic at this concentration, while sucrose, glucose, and sorbitol only reduce the growth rate (Figure4B). Without an osmotic stabilizer, cells were not able to regenerate, indicating that any residual filaments were efficiently removed during purification of the protoplasts (Figure4A). U. maydis protoplasts regenerate into yeast cells and form colonies within three days [17,41]. For T. thlaspeos, we expected filaments to emerge from the protoplast, since we had never observed yeast-cells for this fungus. Furthermore, the slow growth rate of T. thlaspeos cultures suggests a longer

J. Fungi 2021, 7, 38 6 of 14

regeneration time. To confirm our expectation, we described the regeneration process and its timing for T. thlaspeos protoplasts on regeneration medium with 1 M sucrose. After one day, protoplasts turned dark, which is indicative of cell wall regeneration. Three to eight days later, a filament emerges from the protoplast that starts branching after 7–13 days, finally resulting in a micro-colony after 11–18 days (Figure4C). Further proliferation J. Fungi 2021, 7, x FOR PEER REVIEWleads to filamentous colonies after four to five weeks, which are indistinguishable from7 of the 16 original culture.

FigureFigure 3.3. OptimizingOptimizing thethe enzymeenzyme cocktailcocktail for for protoplasting. protoplasting. ( A(A)) Filamentous Filamentous cultures cultures were were harvested harvested and and incubated incubated with with a a combination of 20 mg/mL Glucanex and 10 mg/mL Yatalase, or each enzyme individually, for 30 min. Protoplasting was combination of 20 mg/mL Glucanex and 10 mg/mL Yatalase, or each enzyme individually, for 30 min. Protoplasting was efficient only when both enzymes were applied. (B) The enzymes were diluted to identify the lowest suitable efficient only when both enzymes were applied. (B) The enzymes were diluted to identify the lowest suitable concentration. concentration. The frequency of remaining filaments is inversely proportional to the enzyme concentration. The highest Theefficiency frequency was of obtained remaining with filaments 20 mg/mL is inversely Glucanex proportional and 10 mg/mL to the enzyme Yatalase. concentration. White arrowhead: The highest protoplasts; efficiency black was obtainedarrowhead: with residual 20 mg/mL filaments; Glucanex scale andbar: 1010 mg/mLµm. Yatalase. White arrowhead: protoplasts; black arrowhead: residual filaments; scale bar: 10 µm. Finally, we aimed to decrease the enzyme concentrations to save costs. However, 3.2.lowering Transformation the concentration to half resulted in incomplete digestion of the fungal cell wall after 30 min (Figure 3B). Since it was described earlier that a shorter incubation time is Five antibiotic resistance markers directed against phleomycin, hygromycin, preferable, compared to a low enzyme concentration, regarding protoplast viability nourseothricin, geneticin, and carboxin are routinely used in U. maydis [3]. To develop [39,40], we did not reduce the enzyme concentrations. markers for T. thlaspeos, we tested culture growth on four of these antibiotics. Phleomycin is In contrast to other applications of protoplasts, for genetic manipulation, the a mutagen and therefore was not considered [42]. T. thlaspeos cells were efficiently killed by protoplasts have to be viable and able to regenerate their cell wall. Thus, we next the four antibiotics (Figure S2). Concentration gradients with hygromycin, nourseothricin, investigated the influence of the osmotic stabilizer on regeneration of the protoplasts after and carboxin revealed that T. thlaspeos was more sensitive towards these antibiotics than U. transformation. This allowed us to separate optimization of both steps. Using protoplasts maydis. 10 µg/ hygromycin mL and 50 µg/mL nourseothricin efficiently killed T. thlaspeos generated in the presence of MgSO4, we assessed regeneration media containing different hyphae. This is 20 times and three time less than the standard concentration used for U. osmotic stabilizers such as sucrose, glucose, sorbitol, and KCl (Figure 4). MgSO4 was maydis, respectively. By contrast, cells are less sensitive towards carboxin and remained excluded due to incompatibility with the gelling agent. Maximal regeneration was resistant at 2 µg/mL, the standard concentration used for U. maydis, but were sensitive atobtained 100 µg/mL. with 1 (Figure M sucrose S2). as Carboxin osmotic inhibits stabilizer, the followed mitochondrial by sorbitol succinate and glucose, dehydrogenase while 1 (SDH2),M KCl completely and a point inhibited mutation, regeneration H253L, leads and to fungal a resistant growth form (Figure in U. maydis4A). In [support,43]. The T. thlaspeosthlaspeos SDH2LF1 cell was cultures highly conserved,do not grow with on 82% 1 M amino KCl, acidindicating similarity, that andit is contained toxic at this an arginineconcentration, instead while of the sucrose, histidine glucose, at this position and sorbitol (Figure only S3). reduce This might the growth explain therate reduced (Figure sensitivity.4B). Without an osmotic stabilizer, cells were not able to regenerate, indicating that any residualDue filaments to the high were hygromycin efficiently sensitivity, removed theduring bacterial purification hygromycin-phospho-transferase of the protoplasts (Figure (4A).hpt)[ U.17 maydis] was usedprotoplasts as the regenerate first resistance into yeast marker. cells We and expressed form colonies it as within a fusion three protein days hpt–eGfp[17,41]. For [44 T.] underthlaspeos the, we control expected of T. thlaspeosfilamentsand to emergeU. maydis frompromoters the protoplast, and terminators since we had never observed yeast-cells for this fungus. Furthermore, the slow growth rate of T. (Table S2). Promoter activity was verified in U. maydis. Both PTthsp70 and PTtrps27 were thlaspeos cultures suggests a longer regeneration time. To confirm our expectation, we described the regeneration process and its timing for T. thlaspeos protoplasts on regeneration medium with 1 M sucrose. After one day, protoplasts turned dark, which is indicative of cell wall regeneration. Three to eight days later, a filament emerges from the protoplast that starts branching after 7–13 days, finally resulting in a micro-colony after 11–18 days (Figure 4C). Further proliferation leads to filamentous colonies after four to five weeks, which are indistinguishable from the original culture.

J. Fungi 2021, 7, 38 7 of 14

active in U. maydis, and all five constructs resulted in hygromycin-resistant transformants J. Fungi 2021, 7, x FOR PEER REVIEWwith eGfp-fluorescence (Figure5 and Figure S4), suggesting the constructs are functional8 of 16

and can be used to transform T. thlaspeos.

Figure 4. Optimizing protoplast regeneration. (A) Protoplasts regenerated for 39 days on regeneration medium (YMPG) Figure 4. Optimizing protoplast regeneration. (A) Protoplasts regenerated for 39 days on regeneration medium (YMPG) with either 1 M sorbitol, sucrose, glucose, or KCl. No regeneration was observed without osmotic stabilizer. It was most with either 1 M sorbitol, sucrose, glucose, or KCl. No regeneration was observed without osmotic stabilizer. It was most efficient on sucrose, followed by glucose and sorbitol, while KCl inhibited cell growth. (B) T. thlaspeos LF1 culture plated efficienton YMPG on with sucrose, sorbitol, followed sucrose, by glucose glucose, and KCl, sorbitol, or without while osmotic KCl inhibited stabilizer. cell growth.After 21 (days,B) T. thlaspeosno growthLF1 was culture observed plated for on 1 YMPGM KCl. with Growth sorbitol, rate sucrose, is reduced glucose, for KCl,1 M orsorbitol, without glucose, osmotic and stabilizer. sucrose After compared 21 days, to no absence growth wasof osmotic observed stabilizer. for 1 M KCl. (C) GrowthRegeneration rate is reducedof protoplasts for 1 M on sorbitol, YMPG glucose, with 1 and M sucrosesucrose compared documented to absence microscopically of osmotic stabilizer.for 11–18 (days.C) Regeneration Initially, the of protoplastsprotoplasts onturned YMPG dark with (2 1 Mdays) sucrose before documented new filaments microscopically emerged for(3–8 11–18 days). days. The Initially, filaments the proliferated protoplaststurned (11–18 dark days), (2 days)resulting before in newmacroscopically filaments emerged visible (3–8 colonies days). (28–35 The filaments days). White proliferated arrowhead: (11–18 days),protoplasts; resulting black in macroscopically arrowhead: emerging visible coloniesfilaments; (28–35 scale days).bar: 10 White µm. arrowhead: protoplasts; black arrowhead: emerging filaments; scale bar: 10 µm.

3.2. Transformation Five antibiotic resistance markers directed against phleomycin, hygromycin, nourseothricin, geneticin, and carboxin are routinely used in U. maydis [3]. To develop markers for T. thlaspeos, we tested culture growth on four of these antibiotics. Phleomycin is a mutagen and therefore was not considered [42]. T. thlaspeos cells were efficiently killed by the four antibiotics (Figure S2). Concentration gradients with hygromycin, nourseothricin, and carboxin revealed that T. thlaspeos was more sensitive towards these antibiotics than U. maydis. 10 µg/ hygromycin mL and 50 µg/mL nourseothricin efficiently killed T. thlaspeos hyphae. This is 20 times and three time less than the standard concentration used for U. maydis, respectively. By contrast, cells are less sensitive towards carboxin and remained resistant at 2 µg/mL, the standard concentration used for U. maydis, but were sensitive at 100 µg/mL. (Figure S2). Carboxin inhibits the mitochondrial

J. Fungi 2021, 7, x FOR PEER REVIEW 9 of 16

succinate dehydrogenase (SDH2), and a point mutation, H253L, leads to a resistant form in U. maydis [43]. The T. thlaspeos SDH2 was highly conserved, with 82% amino acid similarity, and contained an arginine instead of the histidine at this position (Figure S3). This might explain the reduced sensitivity. Due to the high hygromycin sensitivity, the bacterial hygromycin-phospho- transferase (hpt) [17] was used as the first resistance marker. We expressed it as a fusion protein hpt–eGfp [44] under the control of T. thlaspeos and U. maydis promoters and J. Fungi 2021, 7, 38 terminators (Table S2). Promoter activity was verified in U. maydis. Both PTthsp70 and PTtrps27 8 of 14 were active in U. maydis, and all five constructs resulted in hygromycin-resistant transformants with eGfp-fluorescence (Figure 5 and Figure S4), suggesting the constructs are functional and can be used to transform T. thlaspeos.

Figure 5. Verification of resistance-reporter constructs in U. maydis. Reporter constructs containing a fusion of hygromycin- Figure 5.phospho-transferaseVerification of resistance-reporter gene (hpt) and theconstructs fluorescent inmarkerU. maydis (egfp .or Reporter mcherry) under constructs the control containing of hsp70 a fusionpromoter of and hygromycin- phospho-transferaseterminator regions gene derived (hpt) andfrom the T. fluorescent thlaspeos genome marker were ( egfptestedor in mcherryU. maydis). underUpon transformation the control ofof thehsp70 linearizedpromoter and terminatorconstruct, regions it derivedrandomly from integrates the T. into thlaspeos the genome.genome Protein were accumulation tested in U.was maydis visualized. Upon by the transformation green/red fluorescence. of the linearized The eGfp expression under the promoter region of T. thlaspeos was stronger than compared to the stably integrated construct, it randomly integrates into the genome. Protein accumulation was visualized by the green/red fluorescence. The construct under the control of a strong, synthetic promoter (Potef). This confirms that the fusion protein is active. In eGfp expression under the promoter region of T. thlaspeos was stronger than compared to the stably integrated construct under the control of a strong, synthetic promoter (Potef ). This confirms that the fusion protein is active. In comparison, mcherry-fluorescence in the strain carrying the Tthsp70 promoter was weaker than the stably integrated construct under the control of the Potef promoter. Scale bar: 10 µm.

First, we needed to define which plasmid to use. Therefore, we transformed an equimolar mixture of five hpt-eGfp plasmids with different promoters (Table S2) into T. thlaspeos LF1 protoplasts generated with the optimized method using the standard U. maydis conditions for transformation [28]. This resulted in a single transformant which had stably integrated the PTthsp70::hpt-egfp:TTthsp70 into the genome (Figure S5, Figure6). Now, transformations using this plasmid regularly result in fluorescent transformants. Based on this successful transformation, a PTthsp70::hpt-mcherry:TTthsp70 construct was generated (Table S2), tested in U. maydis (Figure5), and transformed into T. thlaspeos LF2 (Figure6), showing that compatible mating partners of T. thlaspeos can be tagged with different reporters to follow the mating process. J. Fungi 2021, 7, x FOR PEER REVIEW 10 of 16

comparison, mcherry-fluorescence in the strain carrying the Tthsp70 promoter was weaker than the stably integrated construct under the control of the Potef promoter. Scale bar: 10 µm.

First, we needed to define which plasmid to use. Therefore, we transformed an equimolar mixture of five hpt-eGfp plasmids with different promoters (Table S2) into T. thlaspeos LF1 protoplasts generated with the optimized method using the standard U. maydis conditions for transformation [28]. This resulted in a single transformant which had stably integrated the PTthsp70::hpt-egfp:TTthsp70 into the genome (Figure S5, Figure 6). Now, transformations using this plasmid regularly result in fluorescent transformants. Based on this successful transformation, a PTthsp70::hpt-mcherry:TTthsp70 construct was generated J. Fungi 2021, 7, 38 (Table S2), tested in U. maydis (Figure 5), and transformed into T. thlaspeos LF2 (Figure9 of 6), 14 showing that compatible mating partners of T. thlaspeos can be tagged with different reporters to follow the mating process.

Figure 6. Generation of reporter lines in T. thlaspeos. Reporter constructs containing the active fusion of the hygromycin- Figure 6. Generation of reporter lines in T. thlaspeos. Reporter constructs containing the active fusion of the hygromycin- phospho-transferase gene (hpt) and the fluorescent marker (egfp or mcherry) under the control of the strong hsp70 promoter phospho-transferase gene (hpt) and the fluorescent marker (egfp or mcherry) under the control of the strong hsp70 promoter from T. thlaspeos were transformed into the cultures T. thlaspeos LF1 or LF2, respectively. Fluorescent signals accumulate from T. thlaspeos were transformed into the cultures T. thlaspeos LF1 or LF2, respectively. Fluorescent signals accumulate in the cytosol of all cells. Strains: hpt-egfp: LF1_PTthsp70::hpt-egfp, and hpt-mcherry: LF2_PTthsp70::hpt-mcherry. Scale bar: 10 inµm. the cytosol of all cells. Strains: hpt-egfp: LF1_PTthsp70::hpt-egfp, and hpt-mcherry: LF2_PTthsp70::hpt-mcherry. Scale bar: 10 µm. One key aspect for gene targeted manipulations is efficient homologous recombination.One key aspect To test for whether gene targeted T. thlaspeos manipulations reaches the is same efficient high homologous rates of up to recombina- 50% as U. maydistion. To, we test targeted whether theT. pheromone thlaspeos reaches receptor the gene same pra1 high in ratesthe T. of thlaspeos up to 50% LF1 as backgroundU. maydis, we targeted the pheromone receptor gene pra1 in the T. thlaspeos LF1 background for for deletion. The construct design was based on U. maydis with 1 kb flanking sequences deletion. The construct design was based on U. maydis with 1 kb flanking sequences [27]. [27]. Transformation of the construct resulted in 122 candidates on the transformation Transformation of the construct resulted in 122 candidates on the transformation plates. plates. Reselection of 19 candidates on fresh hygromycin plates led to only nine candidates Reselection of 19 candidates on fresh hygromycin plates led to only nine candidates that that remained resistant. The other candidates were either false positives, or they only remained resistant. The other candidates were either false positives, or they only tran- transiently expressed the resistance protein. These are not interesting for stable siently expressed the resistance protein. These are not interesting for stable integration. integration. In subsequent analysis of the nine candidates, successful deletion of the pra1 In subsequent analysis of the nine candidates, successful deletion of the pra1 locus was locus was confirmed for two transformants (Figure S5), giving a homologous confirmed for two transformants (Figure S5), giving a homologous recombination rate of recombination rate of 22%. 22%. In summary, we have now adapted the protoplast-mediated transformation for the In summary, we have now adapted the protoplast-mediated transformation for the filamentously growing Brassicaceae smut fungus T. thlaspeos. Together, with its ability for filamentously growing Brassicaceae smut fungus T. thlaspeos. Together, with its ability for efficientefficient homologoushomologous recombination, recombination, this this gives gives us a tool us toa study tool plant–microbeto study plant–microbe interactions interactionsof smut fungi of in smut the modelfungi plantin theA. model thaliana plantwith A. two thaliana genetically with two tractable genetically partners. tractable partners. 3.3. Mating of Filaments 3.3. MatingWhen ofT. Filaments thlaspeos teliospores germinate, they give rise to an infectious filament that can directlyWhen T. penetrate thlaspeos theteliospores plant. On germinate, the other they hand, give these rise filaments to an infectious also can filament give rise that to canhaploid directly culture. penetrate Our haploidthe plant. cultures On theT. other thlaspeos hand,LF1 these and filaments LF2 have also compatible can give mating rise to types. They can fuse at the tip and form a new filament [15]. To visualize directional growth of compatible LF1 and LF2 hyphae towards each other during mating, we carried out confrontation experiments. In close proximity, LF1 and LF2 hyphae sense each other, and reorient their growth to meet (Figure7A). In some cases, some hyphae return their growth in direction towards the compatible filament after initial passage. Upon contact, they fuse and result in a new filament (Figure7A,B, Video S1). To prove that fused hyphae really share a common cytoplasm, mating was also observed in cocultivation experiments of compatible strains expressing eGfp and mCherry (Figure7C). After hyphal fusion, eGfp and mCherry fluorescence could be observed in one cytoplasmic segment indicative of plasmogamy. On the other hand, if the pheromone receptor Pra1 is deleted, hyphae of compatible strains grow directly past each other without hyphal fusion (Figure7A,B and Video S2). These findings confirm that the pheromone-receptor system in T. thlaspeos [15] is active and initiates mating. In the future, the generation of nuclei-reporter-strains with J. Fungi 2021, 7, x FOR PEER REVIEW 11 of 16

haploid culture. Our haploid cultures T. thlaspeos LF1 and LF2 have compatible mating types. They can fuse at the tip and form a new filament [15]. To visualize directional growth of compatible LF1 and LF2 hyphae towards each other during mating, we carried out confrontation experiments. In close proximity, LF1 and LF2 hyphae sense each other, and reorient their growth to meet (Figure 7A). In some cases, some hyphae return their growth in direction towards the compatible filament after initial passage. Upon contact, they fuse and result in a new filament (Figure 7A, B, Video S1). To prove that fused hyphae really share a common cytoplasm, mating was also observed in cocultivation experiments of compatible strains expressing eGfp and mCherry (Figure 7C). After hyphal fusion, eGfp J. Fungi 2021, 7, 38 and mCherry fluorescence could be observed in one cytoplasmic segment indicative of 10 of 14 plasmogamy. On the other hand, if the pheromone receptor Pra1 is deleted, hyphae of compatible strains grow directly past each other without hyphal fusion (Figure 7A,B and Video S2). These findings confirm that the pheromone-receptor system in T. thlaspeos [15] NLS-fusion-constructsis active and initiates mating. will allow In the tracking future, the of generation the nuclei of andnuclei-reporter-strains thereby the investigation with of karyogamyNLS-fusion-constructs during mating. will allow tracking of the nuclei and thereby the investigation of karyogamy during mating.

Figure 7. Mating in T. thlaspeos. (A) Top: Confrontation assay of mating partners mfa1/pra1 (LF1) or mfa2/pra2 (LF2). Over the course of 13.5 hours, several mating events of LF1 and LF2 were observed. White arrowheads mark the fusion event. Bottom: Confrontation assay of LF1 pra1∆ and LF2. Over the course of 14 h, hyphae did not mate. Two spots where hyphae could no longer sense each other and cross are marked with black arrow heads. Scale bar: 25 µm. (B,C) Magnification of the mating and crossing events from the white boxes marked in (A). Scale bar: 100 µm. (D) Liquid mating assay with mating partners expressing either cytoplasmic eGfp (LF1-eGfp) or mCherry (LF2-mCherry). Fused hyphae express both eGfp and mCherry and appear yellow in the merged picture (white arrowhead). Scale bar: 25 µm.

4. Discussion When we first set out to work with T. thlaspeos, our aim was to establish a genetically tractable smut fungus in a model host plant such as A. thaliana [15]. An important aim for reaching this goal was genetic manipulation. Here, we show that like other smut fungi, T. thlaspeos is amenable to protoplast-mediated transformation. We have generated a hygromycin resistance cassette, where expression of the hygromycin-phospho-transferase, hpt, is controlled by the T. thlaspeos hsp70 promoter sequence, similar to the cassettes used in U. maydis [3]. Interestingly, promoter sequences seem to be exchangeable between smut fungi, since the T. thlaspeos promoters were active in U. maydis and several groups have J. Fungi 2021, 7, 38 11 of 14

successfully used U. maydis constructs [23,45]. This now enables us to generate reporter strains for a broad range of scientific questions. Most important to establishing a successful protoplasting protocol is the choice of the lytic enzyme(s). The fungal cell wall is a multilayered, chemically complex struc- ture consisting mainly of polysaccharides and varying amounts of lipids, proteins, and polyphosphates [46]. Its composition is not only variable between species [30], but also highly dependent on the culture conditions [47] and morphology [48]. In our case, the combined activity of Yatalase and Glucanex was necessary for efficient digestion of the T. thlaspeos cell wall, although they appear to have overlapping enzymatic properties. Similar additive effects have recently been shown for the ascomycete Hirsutella sinensis [49] and Ag. aegerita [31]; while in Cordyceps militaris, the enzymes mix is less active than Glucanex alone [50]. Hence, during the establishment of conditions for protoplasting, various lytic enzymes and combinations should be tested to reach optimal cell wall degradation [30]. Moreover, commercial manufacturing of enzymes can be stopped, with the broadly used Novozyme 234 being a recent example. Hence, the identification of suitable enzymes can be a reoccurring problem even for established protocols. The second important factor is the osmotic stabilizer, because it depends on the choice of the protoplasting enzyme. For example, the enzymatic activity of Yatalase is inhibited by sorbitol and sucrose. In the 1970s, similar observations were made for helicase [35] and snail enzyme [34]. Protoplasting protocols with Yatalase use inorganic salts as osmotic stabilizer [21,31,37,40,51] and similar to these old reports, for T. thlaspeos, we now use MgSO4 to enable cell wall degradation. Together with other factors influencing the protoplast formation, such as growth conditions of the culture, buffer composition, pH, temperature, or protoplasting time, establishing new transformation protocols quickly becomes a multi-factorial challenge, and testing full-factorial replicates is time-consuming and costly. For T. thlaspeos, we designed pilot studies covering selected combinations in single replicates based on existing transformation protocols, and used the most promising buffer, osmotic stabilizer, and enzyme combination for further optimization. While this approach does not cover all combinations, it allowed us to establish a good transformation protocol with reasonable effort. As a proof-of-principle for our transformation protocol, we investigated the well- characterized smut fungal mating process in T. thlaspeos. In the first step, we generated reporter strains expressing cytosolic eGfp or mCherry to visualize the fusion of hyphae. The resulting filaments express both eGfp and mCherry, indicative of a common cytoplasm, as typical for the dikaryotic smut fungi [52,53]. Next, we looked into dependency on the pheromone receptor. To this end, we generated a deletion mutant of the pheromone receptor pra1 [15] based on the strategy of U. maydis [28]. Notably, homologous recombination also takes place in T. thlaspeos, so we can modify genes in the haploid culture background easily. T. thlaspeos pra1 deletion mutants cannot mate anymore. This finding is especially interesting since it is not yet known whether mating is required for T. thlaspeos to fulfil its life cycle. Infectious filaments emerge directly from germinating T. thlaspeos teliospores. By contrast, teliospore germination of grass smut fungi gives rise to yeast-like sporidia. Subsequent pathogenic development depends on the morphological switch from yeast to filamentous growth brought about by mating [52]. However, the functional conservation of mating genes in T. thlaspeos suggests an evolutionary-conserved, and therefore important, role of mating also in this fungus [15]. This raises several questions. Is mating necessary for the lifecycle of T. thlaspeos? Where and when does mating occur? When do T. thlas- peos hyphae undergo meiosis? Is the filament emerging from the teliospore diploid or dikaryotic? Is the transition to haploid hyphae also occurring naturally in this state of the lifecycle? With the established transformation protocol, we will be able to further address these questions. This will shed light not only onto the mating process of T. thlaspeos, but also on the role of RNA communication in virulence, perennial persistence of the fungus in planta, and nutrition of a smut fungus during biotrophic growth. J. Fungi 2021, 7, 38 12 of 14

5. Conclusions Establishing the genetic manipulation of the Brassicaceae smut fungus T. thlaspeos now allows us to generate reporter strains as well as targeted deletions or modifications of fungal genes. Combined with the fungal colonization of the model plant A. thaliana, we thereby provide a pathosystem, in which both partners have a small, genetically tractable genome for addressing the current and future questions of plant–microbe interactions.

Supplementary Materials: The following are available online at https://www.mdpi.com/2309-6 08X/7/1/38/s1, Figure S1: buffer_comparison, Figure S2: antibiotics, Figure S3: cbx_alignment, Figure S4: transformation_constructs_in_U_maydis, Figure S5: Molecular confirmation of reporter and mutant strains, Table S1: protoplasting_pilot, Table S2: transformation constructs, Video S1: LF1+LF2_confrontation_experiment, Video S2: LF2+LF1-pra1∆_confrontation_experiment, Protocol S1: Plasmid generation, and Protocol S2: Protoplasts and transformation of Thecaphora thlaspeos. Author Contributions: Conceptualization, V.G.; investigation, L.P., K.B., L.G., and P.H.; writing— original draft preparation, L.P. and V.G.; writing—review and editing, V.G.; visualization, L.P.; supervision, V.G.; project administration, V.G.; funding acquisition, V.G. All authors have read and agreed to the published version of the manuscript. Funding: Research in the laboratory of VG was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy (EXC-2048/1, project ID 390686111); the International Research Training NEXTplant (DFG, project ID 391465903/GRK 2466); and the Bioeconomy Science Center (BioSC), which is financially supported by the Ministry of Innovation, Science, and Research within the framework of the NRW Strategieprojekt BioSC (No. 313/323-400-00213). Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: The data presented in this study are available in this manuscript and constructs can be requested from the corresponding author. Acknowledgments: We are grateful to T. Langner, M. Ötztürk, and J. Darenberg for support with cloning of the constructs. R. Kellner helped with discussion and ideas during the development of the transformation protocol. L. Lange provided plates with regenerating protoplasts for microscopy, and K. Vogel supported microscopy. In addition, we would like to thank C. Haag for critically reading the manuscript, M. Feldbrügge for discussions, support, and encouragement throughout the establishment of the method, and P. Künzel for support with preparing the movie files. Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

References 1. Kronstad, J.W. Pathogenesis and sexual development of the smut fungi. In Plant-Microbe Interactions; Stacey, G.K.N.T., Ed.; Springer: Boston, MA, USA, 1997; Volume 1. 2. Feldbrügge, M.; Kämper, J.; Steinberg, G.; Kahmann, R. Regulation of mating and pathogenic development in Ustilago maydis. Curr. Opin. Microbiol. 2004, 7, 666–672. [CrossRef][PubMed] 3. Brachmann, A.; König, J.; Julius, C.; Feldbrügge, M. A reverse genetic approach for generating gene replacement mutants in Ustilago maydis. Mol. Genet. Genom. 2004, 272, 216–226. [CrossRef][PubMed] 4. Banuett, F. Ustilago maydis, the delightful blight. Trends Genet. 1992, 8, 174–180. [CrossRef] 5. Vollmeister, E.; Schipper, K.; Baumann, S.; Haag, C.; Pohlmann, T.; Stock, J.; Feldbrügge, M. Fungal development of the plant pathogen Ustilago maydis. Fems. Microbiol. Rev. 2012, 36, 59–77. [CrossRef][PubMed] 6. Lanver, D.; Tollot, M.; Schweizer, G.; Lo Presti, L.; Reissmann, S.; Ma, L.S.; Schuster, M.; Tanaka, S.; Liang, L.; Ludwig, N.; et al. Ustilago maydis effectors and their impact on virulence. Nat. Rev. Microbiol. 2017, 15, 409–421. [CrossRef][PubMed] 7. Bölker, M.; Urban, M.; Kahmann, R. The a mating type locus of U. maydis specifies cell signaling components. Cell 1992, 68, 441–450. [CrossRef] 8. Gillissen, B.; Bergemann, J.; Sandmann, C.; Schroeer, B.; Bölker, M.; Kahmann, R. A 2-component regulatory system for self non-self recognition in Ustilago maydis. Cell 1992, 68, 647–657. [CrossRef] J. Fungi 2021, 7, 38 13 of 14

9. García-Muse, T.; Steinberg, G.; Pérez-Martín, J. Pheromone-induced G2 arrest in the phytopathogenic fungus Ustilago maydis. Eukaryot. Cell 2003, 2, 494. [CrossRef] 10. Kellner, R.; Vollmeister, E.; Feldbrugge, M.; Begerow, D. Interspecific sex in grass smuts and the genetic diversity of their pheromone-receptor system. PLoS Genet 2011, 7.[CrossRef] 11. Toh, S.S.; Perlin, M.H. Resurgence of less-studied smut fungi as models of phytopathogenesis in the Omics age. Phytopathology 2016, 106, 1244–1254. [CrossRef] 12. Vanky, K.; Lutz, M.; Bauer, R. About the genus Thecaphora (Glomosporiaceae) and its new synonyms. Mycol. Prog. 2008, 7, 31–39. [CrossRef] 13. Andrade, O.; Munoz, G.; Galdames, R.; Duran, P.; Honorato, R. Characterization, in vitro culture, and molecular analysis of Thecaphora solani, the causal agent of potato smut. Phytopathology 2004, 94, 875–882. [CrossRef][PubMed] 14. Conforto, C.; Cazón, I.; Fernández, F.D.; Marinelli, A.; Oddino, C.; Rago, A.M. Molecular sequence data of Thecaphora frezii affecting peanut crops in Argentina. Eur. J. Plant Pathol. 2013, 137, 663–666. [CrossRef] 15. Frantzeskakis, L.; Courville, K.J.; Pluecker, L.; Kellner, R.; Kruse, J.; Brachmann, A.; Feldbrügge, M.; Göhre, V. The plant-dependent life cycle of Thecaphora thlaspeos: A smut fungus adapted to Brassicaceae. Mol. Plant-Microbe Interact. 2017, 30, 271–282. [CrossRef] [PubMed] 16. Courville, K.J.; Frantzeskakis, L.; Gul, S.; Haeger, N.; Kellner, R.; Heßler, N.; Day, B.; Usadel, B.; Gupta, Y.K.; van Esse, H.P.; et al. Smut infection of perennial hosts: The genome and the transcriptome of the Brassicaceae smut fungus Thecaphora thlaspeos reveal functionally conserved and novel effectors. New Phytol. 2019, 222, 1474–1492. [CrossRef][PubMed] 17. Wang, J.; Holden, D.W.; Leong, S.A. Gene transfer system for the phytopathogenic fungus Ustilago maydis. Proc. Natl. Acad. Sci. USA 1988, 85, 865–869. [CrossRef][PubMed] 18. Schirawski, J.; Heinze, B.; Wagenknecht, M.; Kahmann, R. Mating type loci of Sporisorium reilianum: Novel pattern with three a and multiple b specificities. Eukaryot. Cell 2005, 4, 1317–1327. [CrossRef][PubMed] 19. Cervantes-Chavez, J.A.; Ali, S.; Bakkeren, G. Response to environmental stresses, cell-wall integrity, and virulence are orchestrated through the calcineurin pathway in Ustilago hordei. Mol. Plant-Microbe Interact. 2011, 24, 219–232. [CrossRef] 20. Yu, J.; Zhang, Y.; Cui, H.; Hu, P.; Yu, X.; Ye, Z. An efficient genetic manipulation protocol for . FEMS Microbiol. Lett. 2015, 362.[CrossRef] 21. Rabe, F.; Bosch, J.; Stirnberg, A.; Guse, T.; Bauer, L.; Seitner, D.; Rabanal, F.A.; Czedik-Eysenberg, A.; Uhse, S.; Bindics, J.; et al. A complete toolset for the study of Ustilago bromivora and Brachypodium sp as a fungal-temperate grass pathosystem. Elife 2016, 5. [CrossRef] 22. Zuccaro, A.; Basiewicz, M.; Zurawska, M.; Biedenkopf, D.; Kogel, K.-H. Karyotype analysis, genome organization, and stable genetic transformation of the root colonizing fungus Piriformospora indica. Fungal Genet. Biol. 2009, 46, 543–550. [CrossRef] [PubMed] 23. Cheng, Y.; Belzile, F.; Tanguay, P.; Bernier, L.; Bélanger, R. Establishment of a gene transfer system for Pseudozyma flocculosa, an antagonistic fungus of powdery mildew fungi. Mol. Genet. Genom. 2001, 266, 96–102. [CrossRef][PubMed] 24. Lu, S.; Shen, X.; Chen, B. Development of an efficient vector system for gene knock-out and near in-cis gene complementation in the smut fungus. Sci. Rep. 2017, 7, 3113. [CrossRef][PubMed] 25. Ali, S.; Laurie, J.D.; Linning, R.; Cervantes-Chávez, J.A.; Gaudet, D.; Bakkeren, G. An immunity-triggering effector from the barley smut fungus Ustilago hordei resides in an -specific cluster bearing signs of transposable element-assisted evolution. PLoS Pathog. 2014, 10, e1004223. [CrossRef] 26. Ji, L.; Jiang, Z.D.; Liu, Y.; Koh, C.M.; Zhang, L.H. A simplified and efficient method for transformation and gene tagging of Ustilago maydis using frozen cells. Fungal Genet. Biol. 2010, 47, 279–287. [CrossRef][PubMed] 27. Terfrüchte, M.; Joehnk, B.; Fajardo-Somera, R.; Braus, G.H.; Riquelme, M.; Schipper, K.; Feldbrügge, M. Establishing a versatile Golden Gate cloning system for genetic engineering in fungi. Fungal Genet. Biol. 2014, 62, 1–10. [CrossRef][PubMed] 28. Bösch, K.; Frantzeskakis, L.; Vraneš, M.; Kämper, J.; Schipper, K.; Göhre, V. Genetic manipulation of the plant pathogen Ustilago maydis to study fungal biology and plant microbe interactions. J. Vis. Exp. 2016, 115.[CrossRef] 29. Langner, T.; Özturk, M.; Hartmann, S.; Cord-Landwehr, S.; Moerschbacher, B.; Walton, J.D.; Göhre, V. Chitinases Are Essential for Cell Separation in Ustilago maydis. Eukaryot. Cell 2015, 14, 846–857. [CrossRef] 30. Li, D.; Tang, Y.; Lin, J.; Cai, W. Methods for genetic transformation of filamentous fungi. Microb. Cell Factories 2017, 16, 168. [CrossRef] 31. Herzog, R.; Solovyeva, I.; Bölker, M.; Lugones, L.G.; Hennicke, F. Exploring molecular tools for transformation and gene expression in the cultivated edible mushroom Agrocybe aegerita. Mol. Genet. Genom. 2019, 294, 663–677. [CrossRef] 32. de Bekker, C.; Wiebenga, A.; Aguilar, G.; Wösten, H.A.B. An enzyme cocktail for efficient protoplast formation in Aspergillus niger. J. Microbiol. Methods 2009, 76, 305–306. [CrossRef] 33. Ruiz-Díez, B. Strategies for the transformation of filamentous fungi. J. Appl. Microbiol. 2002, 92, 189–195. [CrossRef] 34. Arnold, W.N.; Garrison, R.G. Isolation and characterization of protoplasts from Saccharomyces rouxii. J. Bacteriol. 1979, 137, 1386–1394. [CrossRef] 35. Sietsma, J.H.; De Boer, W.R. Formation and regeneration of protoplasts from Pythium PRL 2142. Microbiology 1973, 74, 211–217. [CrossRef] J. Fungi 2021, 7, 38 14 of 14

36. Tilburn, J.; Scazzocchio, C.; Taylor, G.G.; Zabicky-Zissman, J.H.; Lockington, R.A.; Davies, R.W. Transformation by integration in Aspergillus nidulans. Gene 1983, 26, 205–221. [CrossRef] 37. Hu, Y.; Dietrich, D.; Xu, W.; Patel, A.; Thuss, J.A.J.; Wang, J.; Yin, W.-B.; Qiao, K.; Houk, K.N.; Vederas, J.C.; et al. A carbonate- forming Baeyer-Villiger monooxygenase. Nat. Chem. Biol. 2014, 10, 552–554. [CrossRef] 38. De Vries, O.M.H.; Wessels, J.G.H. Release of protoplasts from Schizophyllum commune by a lytic enzyme preparation from Trichoderma viride. Microbiology 1972, 73, 13–22. [CrossRef] 39. Ferrer, S.; Ramón, D.; Salom, J.; Vicente, E.; Uruburu, F. Protoplasts from Podospora anserina: Isolation, purification, and transformation. Curr. Microbiol. 1985, 12, 301–306. [CrossRef] 40. Zheng, Y.-M.; Lin, F.-L.; Gao, H.; Zou, G.; Zhang, J.-W.; Wang, G.-Q.; Chen, G.-D.; Zhou, Z.-H.; Yao, X.-S.; Hu, D. Development of a versatile and conventional technique for gene disruption in filamentous fungi based on CRISPR-Cas9 technology. Sci. Rep. 2017, 7, 9250. [CrossRef] 41. Eidtmann, A.; Schauz, K. Cryopreservation of protoplasts from sporidia of Ustilago maydis. Mycol. Res. 1992, 96, 318–320. [CrossRef] 42. van Peer, A.F.; de Bekker, C.; Vinck, A.; Wösten, H.A.B.; Lugones, L.G. Phleomycin increases transformation efficiency and promotes single integrations in Schizophyllum commune. Appl. Environ. Microbiol. 2009, 75, 1243–1247. [CrossRef] 43. Keon, J.P.R.; Broomfield, P.L.E.; White, G.A.; Hargreaves, J.A. A mutant form of the succinate dehydrogenase iron—sulphur protein subunit confers resistance to carboxin in the maize smut pathogen Ustilago maydis. Biochem. Soc. Trans. 1994, 22, 234–237. [CrossRef] 44. Lai, J.; Ng, S.K.; Liu, F.F.; Patkar, R.N.; Lu, Y.; Chan, J.R.; Suresh, A.; Naqvi, N.; Jedd, G. Marker fusion tagging, a new method for production of chromosomally encoded fusion proteins. Eukaryot. Cell 2010, 9, 827–830. [CrossRef] 45. Ghareeb, H.; Zhao, Y.; Schirawski, J. Sporisorium reilianum possesses a pool of effector proteins that modulate virulence on maize. Mol. Plant Pathol. 2019, 20, 124–136. [CrossRef] 46. Bartnicki-Garcia, S. Cell wall chemistry, morphogenesis, and of fungi. Annu. Rev. Microbiol. 1968, 22, 87–108. [CrossRef] 47. Aguilar-Uscanga, B.; François, J.M. A study of the yeast cell wall composition and structure in response to growth conditions and mode of cultivation. Lett. Appl. Microbiol. 2003, 37, 268–274. [CrossRef] 48. Gander, J.E. Fungal cell wall glycoproteins and peptido-polysaccharides. Annu. Rev. Microbiol. 1974, 28, 103–120. [CrossRef] 49. Jin, L.-Q.; Xu, Z.-W.; Men, X.-H.; Bo, Z.; Liu, Z.-Q.; Zheng, Y.-G. Enhancement of protoplast preparation and regeneration of Hirsutella sinensis based on process optimization. Biotechnol. Lett. 2020, 42, 2357–2366. [CrossRef] 50. Lou, H.-W.; Ye, Z.-W.; Yu, Y.-H.; Lin, J.-F.; Guo, L.-Q.; Chen, B.-X.; Tang, H.-B.; Wei, T.; Chen, L.-T.; Yun, F. The efficient genetic transformation of Cordyceps militaris by using mononuclear protoplasts. Sci. Hortic. 2019, 243, 307–313. [CrossRef] 51. Kuwano, T.; Shirataki, C.; Itoh, Y. Comparison between polyethylene glycol- and polyethylenimine-mediated transformation of Aspergillus nidulans. Curr. Genet. 2008, 54, 95–103. [CrossRef] 52. Casselton, L.; Feldbrügge, M. Mating and sexual morphogenesis in Basidiomycete fungi. In Cellular and Molecular Biology of Filamentous Fungi; American Society of Microbiology: Washington, DC, USA, 2010. [CrossRef] 53. Morrow, C.A.; Fraser, J.A. Sexual reproduction and dimorphism in the pathogenic basidiomycetes. FEMS Yeast Res. 2009, 9, 161–177. [CrossRef][PubMed]